GUIDELINE
Managing
possible serious
bacterial infection
in young infants
when referral
is not feasible
GUIDELINE
Managing possible serious
bacterial infection in young infants
when referral is not feasible
WHO Library Cataloguing-in-Publication Data
Guideline: Managing possible serious bacterial infection in young infants
when referral is not feasible
I.World Health Organization.
ISBN 978 92 4 150926 8
Subject headings are available from WHO institutional repository
© World Health Organization 2015
All rights reserved. Publications of the World Health Organization are available on the WHO
web site (www.who.int) or can be purchased from WHO Press, World Health Organization,
20 Avenue Appia, 1211 Geneva 27, Switzerland (tel.: +41 22 791 3264; fax: +41 22 791 4857;
e-mail: [email protected]).
Requests for permission to reproduce or translate WHO publications – whether for sale or
for non-commercial distribution – should be addressed to WHO Press through the WHO
website (www.who.int/about/licensing/copyright_form/en/index.html).
The designations employed and the presentation of the material in this publication do
not imply the expression of any opinion whatsoever on the part of the World Health
Organization concerning the legal status of any country, territory, city or area or of its
authorities, or concerning the delimitation of its frontiers or boundaries. Dotted lines on
maps represent approximate border lines for which there may not yet be full agreement.
The mention of specific companies or of certain manufacturers’ products does not imply
that they are endorsed or recommended by the World Health Organization in preference
to others of a similar nature that are not mentioned. Errors and omissions excepted, the
names of proprietary products are distinguished by initial capital letters.
All reasonable precautions have been taken by the World Health Organization to verify
the information contained in this publication. However, the published material is being
distributed without warranty of any kind, either expressed or implied. The responsibility
for the interpretation and use of the material lies with the reader. In no event shall the
World Health Organization be liable for damages arising from its use.
Printed in Switzerland
CONTENTS
Contents
Abbreviations and acronyms
v
Definitions of key terms
vi
Acknowledgementsvii
Executive summary
1
Scope and purpose of the guideline
3
Background4
Objective of guideline
4
Target audience
5
Population of interest
5
Priority questions
6
Outcomes evaluated
8
Methodology9
Evidence retrieval, assessment and synthesis
9
WHO Steering Committee
9
Guideline Development Group
9
Peer Review Group
10
Management of conflicts of interest
10
Grading the quality of evidence
10
Decision-making process
10
Evidence and recommendations
13
13
Review Question 1: Identification of danger signs by
community health worker at home visits
Summary of evidence
13
Considerations for recommendation development
13
Community health worker recommendations
14
Review Question 2: Fast breathing as a single sign of illness
14
Summary of evidence
14
Considerations for recommendation development
15
Fast breathing recommendations
16
iii
GUIDELINE: MANAGING POSSIBLE SERIOUS BACTERIAL INFECTION IN YOUNG INFANTS WHEN REFERRAL IS NOT FEASIBLE
Review Question 3: Clinical severe infection
16
Summary of evidence
17
Considerations for recommendation development
20
Clinical severe infection recommendations
21
Review Question 4: Critical illness
23
Summary of evidence
23
Critical illness recommendations
23
Dissemination, implementation and monitoring of guideline
24
Dissemination
24
Adaptation and implementation
24
Monitoring and evaluation of guideline implementation
25
Implications for future research
26
Plans for updating the guideline
27
References
28
Annexes
Annex 1 Declaration of interests
31
Annex 2 GRADE summary of findings tables
32
iv
CONTENTS
Abbreviations and acronyms
AMR
antimicrobial resistance
CHW
community health worker
CI
confidence interval
GDG
Guideline Development Group
GRADE
Grading of Recommendations, Assessment, Development and
Evaluation
IMintramuscular
IMCI
Integrated Management of Childhood Illness
MCA
Department of Maternal, Newborn, Child and Adolescent Health
PICO Population, Intervention, Control, Outcome
PSBI
possible severe bacterial infection
RD
risk difference
RR
relative risk
UNICEF
United Nations Children’s Fund
USA
United States of America
WHO
World Health Organization
v
GUIDELINE: MANAGING POSSIBLE SERIOUS BACTERIAL INFECTION IN YOUNG INFANTS WHEN REFERRAL IS NOT FEASIBLE
Definitions of key terms
Clinical severe infection: In a young infant (0–59 days old), at least one sign of
severe infection, i.e. movement only when stimulated, not feeding well on
observation, temperature greater than or equal to 38 °C or less than 35.5 °C or
severe chest in-drawing.
Critical illness: In a sick young infant, presence of any of the following signs: uncon­
scious, convulsions, unable to feed at all, apnoea, unable to cry, cyanosis,
bulging fontanelle, major congenital malformations inhibiting oral antibiotic
intake, active bleeding requiring transfusion, surgical conditions needing
hospital referral, persistent vomiting (defined as vomiting following three
attempts to feed the infant within 30 minutes, and the infant vomits after
each attempt).
Fast breathing: In a young infant (0–59 days old), a respiratory rate of greater than or
equal to 60 breaths per minute.
Fast breathing pneumonia: In a young infant (0–59 days old), a respiratory rate of
greater than or equal to 60 breaths per minute as the only sign of possible
infection.
Few events: Fewer than 50 events in the control arm (1).
Neonate: An infant between 0 and 28 days old.
Possible serious bacterial infection: A clinical syndrome used in the Integrated
Management of Childhood Illness package referring to a sick young infant
who requires urgent referral to hospital. The signs include:
• not able to feed since birth or stopped feeding well (confirmed by obser­
vation)
• convulsions
• fast breathing (60 breaths per minute or more) among infants less than 7
days old
• severe chest in-drawing
• fever (38 °C or greater)
• low body temperature (less than 35.5 °C)
• movement only when stimulated or no movement at all.
Sick child: Illness in a child 2–59 months old.
Sick young infant: Illness in an infant 0–59 days old.
vi
ACKNOWLEDGEMENTS
Acknowledgements
The Department of Maternal, Newborn, Child and Adolescent Health of the
World Health Organization gratefully acknowledges the contributions that many
individuals and organizations made to the development of this guideline.
Guideline Development Group
Chair
Maharaj K Bhan
Department of Science and Technology, Government of
India, India
GRADE methodologist
Agustin Conde-Agudelo Department of Obstetrics and Gynaecology, Fundación
Clínica Valle del Lili, Colombia
Members
Caroline Yonaba Okengo HIV and General Paediatric Service, Centre Universitaire
Yalgado Ouedraogo, Burkina Faso
Charu C Garg
Ministry of Health and Family Welfare, India
Dharma S Manandhar
Mother and Infant Research Activities, Nepal
Elizabeth Molyneux
College of Medicine, Malawi
Fred Were
Department of Paediatrics, University of Nairobi, Kenya
Haroon Saloojee
Division of Community Paediatrics, Department
of Paediatrics and Child Health, University of the
Witwatersrand, South Africa
Jonathon Simon
Center for Global Health & Development, Boston
University, USA
Khalid Yunis
American University of Beirut, Lebanon
Kim Mulholland
Murdoch Children’s Research Institute, Royal Children’s
Hospital, Australia
Lulu Bravo
National Institutes of Health, University of the
Philippines, Manila, Philippines
Patricia Hibberd
Department of Pediatrics, Massachusetts General
Hospital for Children, USA
Salim Sadruddin
Save the Children USA, Canada
Syed Abu Jafar Md Musa Ministry of Health and Family Welfare, Bangladesh
Tabish Hazir
ARI Research Cell, Children’s Hospital, Pakistan Institute
of Medical Sciences, Pakistan
vii
GUIDELINE: MANAGING POSSIBLE SERIOUS BACTERIAL INFECTION IN YOUNG INFANTS WHEN REFERRAL IS NOT FEASIBLE
Vinod Paul
WHO Collaborating Centre for Training & Research in
Newborn Care, All India Institute of Medical Sciences,
India
Wally Carlo
Division of Neonatology, University of Alabama at
Birmingham, USA
External Peer Reviewers
Ajay Khera
Ministry of Health and Family Welfare, India
Barbara Stoll
Emory University, USA
Jerome Klein
Boston Medical Center, USA
Joy Lawn
Maternal, Adolescent, Reproductive & Child Health,
London School of Hygiene & Tropical Medicine, United
Kingdom
Linda Wright
National Institutes of Health, USA
Luwei Pearson
UNICEF, Kenya
Mohammad Shahidullah Bangabandhu Sheikh Mujib Medical University,
Bangladesh
Nnenna Ihebuzor
Primary Health Care Systems Development, National
Primary Health Care Development Agency, Nigeria
Sarah Saleem
Department of Community Health Sciences, Aga Khan
University, Pakistan
Trevor Duke
Centre for International Child Health, University of
Melbourne, Department of Paediatrics, Royal Children’s
Hospital, Australia
GRADE systematic reviews and supporting evidence
Simpler antibiotic regimens for managing sick young infants whose families
do not accept referral
Jeeva Shankar
(Team Leader)
Department of Paediatrics, All India Institute of Medical
Sciences, India
Neeraj Gupta
Department of Paediatrics, All India Institute of Medical
Sciences, India
Manisha Mehta
Department of Paediatrics, All India Institute of Medical
Sciences, India
Ramesh Agarwal
Department of Paediatrics, All India Institute of Medical
Sciences, India
Simplified antibiotic regimens for neonates and young infants with fast
breathing
Mike Zangenberg (Team Leader) Department of International Health, Immunology and
Microbiology, Faculty of Health Sciences, University of
Copenhagen, Denmark
Amha Mekasha
Department of Paediatrics and Child Health, Addis Ababa
University, School of Medicine, Ethiopia
viii
ACKNOWLEDGEMENTS
Rakesh Lodha
Department of Paediatrics, All India Institute of Medical
Sciences, India
Sushil K Kabra
Paediatric Pulmonology Division, Department of
Paediatrics, All India Institute of Medical Sciences, India
Home visits by community health workers to improve identification of serious
illness and care seeking from a health facility in newborns and young infants
from low- and middle-income countries
Rakesh Lodha (Team Leader) Department of Paediatrics, All India Institute of Medical
Sciences, India
Amitesh Tripathi
Department of Paediatrics, All India Institute of Medical
Sciences, India
SK Kabra
Department of Paediatrics, All India Institute of Medical
Sciences, India
HPS Sachdev
Sitaram Bhartiya Institute of Medical Sciences, India
Feasible, acceptable and cost-effective models of service delivery for neonates
and young infants aged 0–59 days with possible serious bacterial infection:
a systematic review
Jai K Das
(Team Leader)
Division of Women and Child Health, Aga Khan
University, Pakistan
Zohra S Lassi
Division of Women and Child Health, Aga Khan
University, Pakistan
Mehar Hoda
Division of Women and Child Health, Aga Khan
University, Pakistan
Rehana A Salam
Division of Women and Child Health, Aga Khan
University, Pakistan
Cost and cost-effectiveness of the management of possible serious bacterial
infection in young infants: a systematic review
Harish Nair (Team Leader)
Centre for Population Health Sciences, University of
Edinburgh, United Kingdom
Harry Campbell
Centre for Population Health Sciences, University of
Edinburgh, United Kingdom
Shanshan Zhang
Centre for Population Health Sciences, University of
Edinburgh, United Kingdom
WHO Steering Committee
Shamim Qazi
Department of Maternal, Newborn, Child and Adolescent
Health, Switzerland
Rajiv Bahl
Department of Maternal, Newborn, Child and Adolescent
Health, Switzerland
ix
GUIDELINE: MANAGING POSSIBLE SERIOUS BACTERIAL INFECTION IN YOUNG INFANTS WHEN REFERRAL IS NOT FEASIBLE
Other WHO staff and consultants
Samira Aboubaker
Department of Maternal, Newborn, Child and Adolescent
Health, Switzerland
Andrew Mbewe
Regional Office for Africa, Congo
Rajesh Mehta
Regional Office for South-East Asia, India
Lulu Muhe
Department of Maternal, Newborn, Child and Adolescent
Health, Switzerland
Nalini Singh
Pandemic and Epidemic Diseases, Switzerland
Sachiyo Yoshida
Department of Maternal, Newborn, Child and Adolescent
Health, Switzerland
Severin von Xylander
Department of Maternal, Newborn, Child and Adolescent
Health, Switzerland
Funding and acknowledgement
A grant from the United States Agency for International Development supported
the development of this publication. WHO is grateful to all the individuals and
institutions that contributed time and made other in-kind contributions to the
guideline development process. We would like to acknowledge the contribution
made by Dr Mathew Fox, Boston University, USA; Dr Lulu Muhe, Switzerland; and
Dr Peggy Henderson, Switzerland, who assisted with the development of this
document.
x
EXECUTIVE SUMMARY
Executive summary
Every year, about 2.8 million children die in the first month of life, with 98% of these
deaths occurring in developing countries. The current WHO recommendation for
management of infections in neonates and young infants (0–59 days old) is referral
for hospital treatment with a seven to 10 day course of a combination of two injectable
antibiotics – penicillin or ampicillin plus gentamicin. However, existing evidence
demonstrates that in resource-limited settings many young infants with signs of
possible serious bacterial infection (PSBI) do not receive the recommended inpatient
treatment because such treatment is not accessible, acceptable or affordable to
families. While increasing hospital-based treatment by improving availability and
access is imperative, providing effective treatment for young infants with severe
infection at first-level health facilities when families do not accept or cannot access
referral would increase access to potentially lifesaving care for these infants.
Although previously there has been little evidence to evaluate the safety and efficacy
of providing care to young infants with PSBI at lower level facilities, a body of research
has been conducted over the past decade to inform the creation of evidence-based
guidelines. Evaluating data from recently completed studies, it is now possible to
develop global clinical and programmatic guidance on management of PSBI where
referral for treatment is not feasible.
This guideline, developed by a panel of international experts and informed by
a thorough review of existing evidence, provides recommendations on the use
of antibiotics for neonates and young infants (0–59 days old) with PSBI in order to
reduce young infant mortality rates. This guideline is intended for use in resourcelimited settings in situations when families do not accept or cannot access referral.
It seeks to provide programmatic guidance on the role of community health workers
(CHWs) and home visits in identifying signs of serious infections in neonates and
young infants. It also seeks to provide clinical guidance on the simplest antibiotic
regimens that are both safe and effective for outpatient treatment of clinical severe
infections and fast breathing (pneumonia) in children 0–59 days old.
This guideline will not replace the WHO-recommended inpatient management as
the preferred treatment option for young infants who have clinical severe infection
or critical illness. Close follow-up is essential for young infants managed on an
outpatient basis where referral is not possible.
To develop these recommendations, a WHO Steering Committee and an
18-member Guideline Development Group (GDG) of experts was convened. GDG
members each declared their interests, and no conflicts of interest were identified.
The group developed a series of priority questions, and WHO commissioned
independent institutions to conduct systematic reviews for each question. Based
on these reviews, the WHO Steering Committee developed an initial set of draft
recommendations. Members of the GDG then reviewed and evaluated the quality of
the evidence identified through the systematic reviews using the GRADE (Grading of
Recommendations, Assessment, Development and Evaluation) methodology (www.
1
GUIDELINE: MANAGING POSSIBLE SERIOUS BACTERIAL INFECTION IN YOUNG INFANTS WHEN REFERRAL IS NOT FEASIBLE
gradeworkinggroup.org) and revised and finalized the guideline recommendations.
The final recommendations, which were approved by the WHO Guidelines Review
Committee, appear in the summary of recommendations below.
The target audience for this guideline includes: 1) national policy-makers in health
ministries; 2) programme managers working in child health, essential drugs and
health worker training; 3) health care providers and clinicians managing sick
children at various levels of health care, including public and private practitioners;
and 4) development partners providing financial and/or technical support for child
health programmes.
2015 WHO Recommendations on managing possible serious bacterial infection in young
infants 0–59 days old when families do not accept or cannot access referral care
No.
1
2
3
4
5
Strength of
recommendation
Quality of
evidence
Strong
Moderate
Strong
Low
Strong
Low
Option 1: Intramuscular gentamicin 5–7.5 mg/kg (for low-birth-weight
infants gentamicin 3–4 mg/kg) once daily for seven days and twice daily oral
amoxicillin, 50 mg/kg per dose for seven days. Close follow-up is essential.
Strong
Moderate
Option 2: Intramuscular gentamicin 5–7.5 mg/kg (for low-birth-weight infants
gentamicin 3–4 mg/kg) once daily for two days and twice daily oral amoxicillin,
50 mg/kg per dose for seven days. Close follow-up is essential. A careful
assessment on day 4 is mandatory.
Strong
Low
Strong
Very low
Recommendation
Community health workers and home visits for postnatal care
At home visits made as part of postnatal care (2), community health workers
should counsel families on recognition of danger signs, assess young infants
for danger signs of illness and promote appropriate care seeking.a
Infants 0–6 days with fast breathing as the only sign of illness
Young infants 0–6 days old with fast breathing as the only sign of illness should
be referred to hospital. If families do not accept or cannot access referral care,
these infants should be treated with oral amoxicillin, 50 mg/kg per dose twice
daily for seven days, by an appropriately trained health worker.
Infants 7–59 days with fast breathing as the only sign of illness
Young infants 7–59 days old with fast breathing as the only sign of illness
should be treated with oral amoxicillin, 50 mg/kg per dose twice daily for
seven days, by an appropriately trained health worker. These infants do not
need referral.
Young infants 0–59 days old with clinical severe infection
Young infants 0–59 days old with clinical severe infection whose families do
not accept or cannot access referral care should be managed in outpatient
settings by an appropriately trained health worker with one of the following
regimens:b
Young infants 0–59 days old with critical illness
Young infants 0–59 days old who have any sign of critical illness (at
presentation or developed during treatment of clinical severe infection)
should be hospitalized after pre-referral treatment (3).c
If birth is in a health facility, mothers and newborns should receive care in the facility for at least 24 hours. If at home, first postnatal contact
should be as early as possible within 24 hours. At least three additional postnatal contacts are recommended for all mothers and newborns on
day 3 (48–72 hours), between days seven to 14, and six weeks after birth.
b
Option 1 is the preferred option, but where the health system does not allow this to be implemented, option 2 could be considered. The GDG
felt that option 2 likely would be easier to deliver, have more equitable access, have higher adherence, be more affordable and have similar
effectiveness. It is expected that individual countries will adapt the recommendations to suit the local social, cultural and economic contexts.
Countries are encouraged to hold key stakeholder discussions to inform the decision-making on use and introduction of the recommendations
into national programmes.
c
Give first dose of both ampicillin (50 mg/kg per dose) or benzyl penicillin (50 000 units/kg per dose) and gentamicin (5–7.5 mg/kg per dose)
intramuscularly.
a
2
SCOPE AND PURPOSE OF THE GUIDELINE
Scope and purpose
of the guideline
This guideline, developed by a panel of international experts and informed by a
thorough review of existing evidence, contains a number of recommendations on
the use of antibiotics for neonates (0–28 days old) and young infants (0–59 days old)
with PSBI in order to reduce young infant mortality rates. The guideline is intended
for use in resource-limited settings in situations when families do not accept or
cannot access referral care. The goal of the guideline is to provide clinical guidance
on the simplest antibiotic regimens that are both safe and effective for outpatient
treatment of clinical severe infections and fast breathing (pneumonia) in children
0–59 days old. In addition, the guideline seeks to provide programmatic guidance
on the role of CHWs and home visits in identifying signs of serious infections in
neonates and young infants.
This guideline will help health care providers make appropriate management
decisions about sick young infants whose families cannot access referral care. They
will also guide national policy-makers in health ministries, programme managers,
and development partners and will inform revisions to current WHO training and
reference materials such as the Integrated Management of Childhood Illness (IMCI) Chart
Booklet (3) and the Pocket Book of Hospital Care for Children: guidelines for the management
of common illnesses (4).
This guideline is intended for resource-limited settings in the context of primary
health care only. It is intended for the use of professionally trained health workers
only and not for lay CHWs (with the exception of recommendations specific to
CHWs). This guideline requires appropriate training of health care workers,
supplying them with necessary equipment, job aids and medicines and providing
adequate supervision and oversight. Monitoring of the programme will be needed
for ensuring success and identifying and documenting adverse events of medicines
(e.g. gentamicin toxicity), and community perceptions should be closely monitored.
In addition surveillance for antimicrobial resistance (AMR) should be strengthened.
This guideline will not replace the WHO-recommended inpatient management as
the preferred treatment option for young infants who have signs of severe infection
(4). Close follow-up is essential for young infants managed on an outpatient basis
where referral is not possible.
3
BACKGROUND
Background
Every year, about 2.8 million children die in the first month of life, with 98% of these
deaths occurring in developing countries (5). Neonatal infections, including sepsis
and meningitis, are estimated to cause over 420 000 deaths each year, with 136 000
attributed to pneumonia (6). The current WHO recommendation for management of
infections in neonates (0–28 days old) and young infants (0–59 days old) is referral for
hospital treatment with at least a seven-day course of a combination of two injectable
antibiotics – benzylpenicillin or ampicillin plus gentamicin (3,4). However, existing
evidence demonstrates that in resource-limited settings many young infants with
signs of severe infection do not receive the recommended inpatient treatment
because such treatment is not accessible, acceptable or affordable to families (7–
14). Non-compliance with referral for hospitalization means that these infants
receive no treatment, resulting in unnecessary, potentially preventable infectionrelated newborn deaths. While increasing hospital-based treatment by improving
availability and access is imperative, if effective treatment for young infants with
severe infection could be provided at first level health facilities or at the community
level when families do not accept or cannot access referral care, many more infants
could access potentially lifesaving care.
Research studies in India and Bangladesh have shown that it is possible to treat
neonates 0–28 days old with signs of severe infection with injectable gentamicin
in combination with oral cotrimoxazole or injectable procaine penicillin in the
community when referral is not accepted by families (8,9). A subsequent study from
Pakistan showed that the efficacy of a procaine penicillin-gentamicin combination
was much higher than that of cotrimoxazole-gentamicin (10). No data were available
from Africa. Whereas the Asian studies provided proof of principle, the proposed
treatment was difficult to deliver in programmatic settings. Simpler regimens
combining oral and intramuscular antibiotics or involving a switch from injectable
to oral antibiotics after the first two to three days of treatment might be easier to
deliver at outpatient level in resource-limited health systems.
Objective of guideline
While referral for hospital care continues to be the preferred method for the
management of neonates and young infants with PSBI, a guideline for first level
health facilities or community-based management of common life-threatening
infections is needed to facilitate timely case management in cases where referral
is either not accepted or not accessible. While previously there was little evidence
to evaluate the safety and efficacy of providing care to young infants with PSBI at
lower level facilities, a body of research has been conducted over the past decade
to inform the creation of an evidence-based guideline. This includes a group of
research projects supported by the Bill & Melinda Gates Foundation and the United
States Agency for International Development, which Save the Children’s Saving
Newborn Lives programme and WHO conducted in five countries in South Asia
4
BACKGROUND
and sub-Saharan Africa. As these studies have now been completed, a review was
undertaken to develop global clinical and programmatic guidance on management
of PSBI where referral for treatment is not possible or not accessible.
Target audience
The target audience for this guideline includes: 1) national policy-makers in health
ministries; 2) programme managers working in child health, essential drugs and
health worker training; 3) health care providers and clinicians managing sick
children at various levels of health care, including public and private practitioners;
and 4) development partners providing financial and/or technical support for child
health programmes.
Population of interest
This guideline focuses on management of neonates (0–28 days old) and young infants
(0–59 days old) in resource-limited settings who have clinical signs of PSBI and whose
families do not accept or cannot access referral care for treatment. Clinical signs of
PSBI are defined as:
• fast breathing (respiratory rate ≥ 60 breaths/minute)
• severe chest in-drawing
• fever (temperature ≥ 38 °C)
• hypothermia (temperature < 35.5 °C)
• no movement at all or movement only on stimulation
• feeding poorly or not feeding at all
• convulsions.
As previously noted, hospitalization remains the first priority for care for these
infants. This guideline only applies in cases where hospitalization is not accepted or not
accessible.
Justification for this age category
The WHO Global Acute Respiratory Infection Control Programme, launched in the
late 1980s, used the age categories of up to 2 months and 2 months up to 59 months
for management of children with acute respiratory infection and pneumonia. The
signs and management of infants with pneumonia 29–59 days of age were similar to
infants 0–28 days, as opposed to children 2–59 months of age (15). IMCI, which was
first launched in 1997, used the same age cut-offs because the signs and management
of PSBI in infants 29-59 days were similar to infants 0–28 days of age, as opposed to
the signs and management in children 2–59 months of age (3).
This guideline provides recommendations for outpatient treatment for two subgroups
of young infants with PSBI, based on increasing clinical severity. The subgroups are
defined as:
• Fast breathing pneumonia: A young infant (0–59 days old) with a respiratory rate
of ≥ 60 breaths per minute as the only sign of possible infection.
• Clinical severe infection: A young infant (0–59 days old) with at least one sign
of severe infection (i.e. movement only when stimulated, not feeding well on
observation, temperature ≥ 38 °C or < 35.5 °C or severe chest in-drawing).
5
GUIDELINE: MANAGING POSSIBLE SERIOUS BACTERIAL INFECTION IN YOUNG INFANTS WHEN REFERRAL IS NOT FEASIBLE
Priority questions
In order to identify and evaluate the evidence for outpatient treatment, the WHO
Steering Committee developed four main priority questions on the safety and
effectiveness of simpler antibiotic regimens for treatment of PSBI in young infants.
Each was described in terms of its characteristics: population, intervention, control
and outcome (PICO). Simpler antibiotic regimens were defined as regimens with fewer
injections than the reference treatment of intramuscular penicillin and gentamicin
for at least seven days (i.e. 14 injections). This could be one antibiotic injection
for seven days with or without an oral antibiotic (i.e. seven injections), initiating
treatment with intramuscular treatment and switching to an oral antibiotic after
two to three days (i.e. two to six injections) or oral antibiotic treatment alone (i.e. no
injections).
To summarize the evidence, the following questions about infants whose families
do not accept or cannot access referral care for treatment were formulated as the
basis for systematic reviews:
1. Among all neonates 0–28 days old, can home visits by CHWs compared to no home
visits successfully identify newborns with serious illness and improve care
seeking from a health facility?
a. Population: Neonates 0–28 days old.
b. Intervention: Home visits by CHWs.
c. Control: No home visits by CHWs.
d. Outcome: Identification of newborns with serious illness and improvement in
care seeking from a health facility.
2. Among neonates (0–6 days old) and young infants (7–59 days old) presenting with fast
breathing as a single sign of PSBI, are simpler antibiotic regimens, delivered at
out­­patient and/or community level, as effective as a combination of injectable
penicillin and gentamicin for at least seven days as measured by rates of mortality
and clinical deterioration (development of signs of severe infection or critical
illness), within two weeks of starting treatment?
a. Population: Neonates (0–6 days old) and young infants (7–59 days old)
presenting with fast breathing as a single sign of PSBI.
b. Intervention: Simpler antibiotic regimens delivered at outpatient and/or
community level.
c. Control: A combination of injectable penicillin and gentamicin for at least
seven days.
d. Outcome: Mortality and clinical deterioration within two weeks of starting
treatment.
3. Among neonates and young infants (0–59 days old) with suspected/confirmed clinical
severe infection, can simpler antibiotic regimens, delivered at outpatient and/
or community level, be as effective as a combination of injectable penicillin
and gentamicin given for at least seven days in achieving comparable rates of
mortality, clinical deterioration (i.e. development of any sign of critical illness) and
persistence of signs of severe infection within two weeks of starting treatment?
6
BACKGROUND
a. Population: Neonates and young infants (0–59 days old) with suspected/
confirmed clinical severe infection.
b. Intervention: Simpler antibiotic regimens delivered at outpatient and/or
commu­nity level.
c. Control: A combination of injectable penicillin and gentamicin given for at
least seven days.
d. Outcome: Mortality, clinical deterioration and persistence of signs of severe
infection within two weeks of starting treatment.
4. Among neonates and young infants (0–59 days old) with signs of critical illness, are simpler
antibiotic regimens delivered at outpatient and/or community level, as effective
as a combination of injectable penicillin and gentamicin for at least seven days as
measured by rates of mortality, clinical deterioration (i.e. development of a new
sign of critical illness) and persistence of signs of severe infection, within two
weeks of starting treatment?
a. Population: Neonates and young infants (0–59 days old) with signs of critical
illness.
b. Intervention: Simpler antibiotic regimens delivered at outpatient and/or
community level.
c. Control: A combination of injectable penicillin and gentamicin for at least
seven days.
d. Outcome: Mortality, clinical deterioration and persistence of signs of severe
infection, within two weeks of starting treatment.
Systematic reviews addressing each of the above-mentioned priority questions were
conducted, which informed the evidence presented in this document. In addition two
other priority questions for systematic reviews were formulated to provide context
and additional information on the possible treatments for outpatient care among
infants whose families do not accept or cannot access referral care for treatment:
1. Among neonates and young infants (0–59 days old) who have signs of PSBI, who are
managed with simpler antibiotic regimens delivered at outpatient and/or com­
munity level, what are the total costs/cost-effectiveness of treatment with simpler
effective antibiotic regimens (defined as having equivalent or better reductions
in mortality, clinical deterioration, non-recovery and relapse), compared to a
combination of injectable penicillin and gentamicin for at least seven days?
a. Population: Neonates (0–28 days old) and young infants (0–59 days old) who
have signs of PSBI.
b. Intervention: Management with simpler antibiotic regimens delivered at out­
patient and/or community level.
c. Control: Combination of injectable penicillin and gentamicin for at least seven
days.
d. Outcome: Total costs/cost-effectiveness of treatment with simpler effective
antibiotic regimens.
7
GUIDELINE: MANAGING POSSIBLE SERIOUS BACTERIAL INFECTION IN YOUNG INFANTS WHEN REFERRAL IS NOT FEASIBLE
2. Among neonates and young infants (0–59 days old) who have signs of PSBI, who are
managed with simpler antibiotic regimens delivered at outpatient and/or
community level, compared with a combination of injectable penicillin and
gentamicin for at least seven days, what are the feasible, acceptable, effective
and cost-effective models of service delivery1 and what are the health system
requirements for all these models?
a. Population: Neonates and young infants (0–59 days old) who have signs of
PSBI.
b. Intervention: Management with simpler antibiotic regimens delivered at out­
patient and/or community level.
c. Control: A combination of injectable penicillin and gentamicin for at least
seven days.
d. Outcome: Feasibility, acceptability, effectiveness and cost-effectiveness of
models of service delivery.
The results of these last two systematic reviews are not presented directly but rather
informed the discussions that led to the recommendations that follow.
Outcomes evaluated
For each priority question two critical and two important outcomes were evaluated.
The two critical outcomes were: 1) death within two weeks of starting treatment
for PSBI; and 2) treatment failure by day 8 after starting treatment, defined as
clinical deterioration after starting treatment for PSBI including death, need for
hospitalization/referral, need to change antibiotics, new signs of clinical infection
or persistence of signs of illness. The two important outcomes were: 1) clinical
relapse (defined as emergence of any sign of infection within two weeks after the
disappearance of all signs of clinical PSBI); and 2) adherence to treatment (defined as
compliance or adherence to treatment as specified in the study methods). Summaries
of all outcomes, but only the GRADE tables for the critical outcomes, are presented,
as these were the focus of the GDG.
1
These could include: 1) self-referral, with treatment in outpatient health facilities; 2) use of
CHWs to identify sick young infants and refer them to outpatient clinics for care; and 3) use
of CHWs to identify and treat young infants at the community level.
8
METHODOLOGY
Methodology
The WHO Steering Committee organized a meeting in Geneva on 7–8 October 2013 to
discuss the results of the recently concluded research. The participants were WHO
Staff, WHO Steering Committee members and GDG members (described below).
The GDG decided there was adequate evidence to update the WHO guideline on
management of PSBIs in young infants where families do not accept or cannot access
referral. The GDG then developed the priority questions (as listed above) and WHO
commissioned independent institutions to conduct systematic reviews of each
question.
Evidence retrieval, assessment and synthesis
The evidence retrieval process for the priority questions followed the standard
outlined in the WHO Handbook for Guideline Development (16). A list of reviewers is
provided in the Acknowledgements. A protocol for each systematic review was
developed and included the search terms and strategy, and the PICOs used to
define the inclusion and exclusion criteria. The detailed search strategy for each
priority question was agreed upon after a series of discussions with the Steering
Committee and lead investigators of each systematic review. Each review includes
a flow diagram showing the numbers of studies excluded and included. Medline
and EMBASE databases were used to identify peer-reviewed publications. The WHO
International Clinical Trials Registry Platform, the Cochrane Central Register of
Controlled Trials, the International Standard Randomised Controlled Trial Number
Register, and ClinicalTrials.gov were searched for on-going studies. The quality of
the evidence for each priority question was assessed using the GRADE methodology
(www.gradeworkinggroup.org). The quality of the evidence for treatment interven­
tions was graded as high, moderate, low or very low based on the definitions in
the GRADE guide (17). The GRADE tables were prepared using the GRADE profiler
software, where appropriate. The reviews are available on file and will be published.
Based on these reviews, the WHO Steering Committee proposed an initial set of draft
recommendations.
WHO Steering Committee
A Steering Committee, with members from the Department of Maternal, Newborn,
Child and Adolescent Health (MCA), oversaw the guideline review process. WHO
staff are listed in the Acknowledgements.
Guideline Development Group
WHO convened an 18-member GDG consisting of internationally recognized experts.
A list of members is provided in the Acknowledgements. Members of the group
were tasked with reviewing and evaluating the quality of the evidence identified
through the systematic reviews using the GRADE methodology (described below)
and revising and finalizing the guideline recommendations.
9
GUIDELINE: MANAGING POSSIBLE SERIOUS BACTERIAL INFECTION IN YOUNG INFANTS WHEN REFERRAL IS NOT FEASIBLE
Peer Review Group
The duties of the Peer Review Group were to review the recommendations and
supporting documentation developed by the GDG. The list of members, from various
countries and disciplines, with their affiliations is provided in the Acknowledgements.
Management of conflicts of interest
All of the members of the GDG were required to sign and submit a Declaration of
Interests prior to their participation in the meetings. The WHO Steering Group
reviewed the Declarations prior to the GDG meeting to determine whether a conflict
existed that might have precluded or limited anyone’s participation. While no
conflicts of interest were declared requiring any action, the interests declared are
given in Annex I.
Grading the quality of evidence
The GRADE methodology was used by the GDG to evaluate the quality of the
evidence. This methodology is a widely-used approach for separating the quality of
evidence from the strength of recommendations (17). The Cochrane Collaboration
and WHO (16) have adopted it as standard for developing systematic reviews and
recommendations. GRADE tables summarize details about the studies reviewed,
including study outcomes, limitations, possible inconsistency, indirectness,
imprecision and other factors that might impact judgements on the quality of
evidence. GDG members then used the information to define the overall quality of
evidence as very low, low, moderate or high as defined in Table 1.
Table 1. Definition of quality of evidence using the GRADE methodology
Quality
Definition
High
Further research is unlikely to change confidence in the estimates of the effect
Moderate
Further research is likely to have an important impact on confidence in the estimate of the effect and may
change the estimate
Low
Further research is very likely to have an important impact on the confidence of the effect and is likely to
change the estimate
Very low
Any estimate of effect is very uncertain
Decision-making process
The WHO Steering Committee in Geneva convened a GDG meeting from 19–21
November 2014. Each member of the GDG was provided with a copy of the com­
missioned systematic reviews.
For each draft recommendation, the WHO Steering Committee presented a synthesis
of the evidence, the GRADE tables and a draft recommendation. Decision-making
tables were also prepared including benefits and risks of the interventions from a
public health perspective; values, preferences and acceptability to programme
managers and policy-makers and health care providers; and feasibility of implemen­
ting any recommendations (including resources needed, focusing on national
programmes in resource-limited or other settings). Each member was required to
review the material and independently complete a written form asking them to
10
METHODOLOGY
comment on and suggest revisions to the proposed guidance and decision-making
tables. The form also required them to rate, for each recommendation: the overall
quality of evidence using the GRADE methodology (independent of the rating made
in the synthesis of the evidence), the balance of benefits versus harms, values that
should be considered in making a recommendation, and applicability of any proposed
recommendations to the populations they are intended to benefit. Finally they were
asked to give their assessment of what the strength of each recommendation should
be based on the criteria provided in Table 2.
Table 2. Assessment criteria for the strength of recommendations
Strength of recommendation
Rationale
Strong
The GDG is confident that the desirable effects of adherence to the
recommendation outweigh the undesirable effects.
Weak
The GDG concludes that the desirable effects of adherence to a recommendation
probably outweigh the undesirable effects. However, the recommendation is only
applicable to a specific group, population or setting OR where new evidence may
result in changing the balance of risk to benefit OR where the benefits may not
warrant the cost or resource requirements in all settings.
No recommendation
Further research is required before any recommendation can be made.
The GDG then used a consensus building process to finalize the recommendations.
Once each participant had completed the form detailing their opinions and proposed
language, a summary of the opinions of the entire group was created by members of
the WHO Steering Committee. This was presented to the GDG members in order to
gauge consensus and where there were great differences. At this point the chair led
a discussion on each recommendation, and the group edited the recommendations.
Finally the chair of the GDG made an attempt to reach consensus among GDG
members on the language for the recommendation, the quality of evidence and the
strength of recommendation.
WHO staff did not express any personal opinions on the data, in the discussions or
in the decisions on language, strength of recommendations or quality of evidence.
At all stages of the meeting they articulated in detail the principles and guidelines of
the WHO decision-making process.
In all cases except one, the committee reached unanimous agreement on the recom­
mendations as revised by the GDG. The exception was for Recommendation 4 option
2 allowing for treatment of clinical severe infection in children 0–59 days old with
intramuscular gentamicin once daily 5–7.5 mg/kg per day for two days and twice daily
oral amoxicillin, 50 mg/kg per dose for seven days. One member of the GDG did not
agree with this recommendation. The group decided that if consensus could not be
reached, a two thirds majority would be required to put forward a recommendation.
Additional efforts to address the concerns raised1 were not successful, but as 17 of
the 18 members agreed the recommendation was put forward.
1
The dissenting member expressed his concerns in written form to the WHO Steering
Committee. The GDG chair formulated a written response to the concerns raised. In order
to ensure these concerns were not ignored, both documents were sent to each remaining
member of the committee with a request to comment on whether they would like to change
their decision or make changes to the wording of the recommendation. Each responded that
they still agreed with the recommendation as drafted in the November 2014 meeting, and
therefore the GDG decided not to change it.
11
GUIDELINE: MANAGING POSSIBLE SERIOUS BACTERIAL INFECTION IN YOUNG INFANTS WHEN REFERRAL IS NOT FEASIBLE
The WHO Steering Committee then sought external peer review of the recom­
mendations. The draft recommendations were sent to an external Peer Review
Group for input on proportionality, format, clarity, consistency, level of detail, gaps,
errors, implementation considerations and other general comments. The external
peer reviewers made several suggestions to improve the document. The Steering
Committee reviewed all suggestions and incorporated the reviewers’ comments
as appropriate, such as improving the evidence base for the fast breathing
recommendation for young infants 0–6 days old, improving the sub-section on
monitoring and evaluation, improving the general writing of text and mode of
presen­tation and updating the list of references. The final version was approved by
the GDG.
12
EVIDENCE AND RECOMMENDATIONS
Evidence and
recommendations
Review Question 1: Identification of danger signs by community health
worker at home visits
The systematic review was based on the priority question: among all neonates 0–28
days old (Population), can home visits by CHWs (Intervention) compared to no home
visits (Control) successfully identify young infants with serious illness and improve
care seeking from a health facility (Outcome)?
Summary of evidence
Six randomized controlled trials (14,18–22) compared home visits by a CHW to
identify young infants with serious illness (Intervention) to no home visits (Control)
in children 0–59 days old (Population). The six studies, which are described in Grade
Table 1 (Annex 2), randomized a total of 4760 infants to an intervention and 4398 to
a control group. The intervention was evaluated based on improved care seeking
at health facilities (Outcome). After pooling the data from the six trials, there was
significant improvement in care seeking in the intervention arm compared to the
control arm (relative risk [RR]: 1.35; 95% confidence interval [CI]: 1.15–1.58).
Considerations for recommendation development
Balance of benefits versus harms: The GDG noted that at postnatal home visits,
identification of danger signs in young infants by CHWs is beneficial because it
can increase early health care seeking. In addition, the specificity of diagnosis of
serious illness by CHWs appears to be high (i.e. few false positives), and therefore the
likelihood of unnecessary referrals is low. However, close supervision and support
will still be needed to avoid missing infants with danger signs.
Values and preferences: The GDG noted that home visits for neonates 0–28 days
old are already part of WHO and country strategies, and therefore it is a missed
opportunity if CHWs are not trained and supported to identify PSBI and promote
care seeking.
Feasibility (including resource use considerations): The GDG noted that postnatal
care of newborns1 is already recommended by WHO (2) but some additional costs
will be required under this recommendation for training CHWs to identify sick
young infants. The GDG felt those costs were worth the benefit. In addition, WHO
training materials for CHWs will need to be modified.
1
If birth is in the health facility, mothers and newborns should receive care in the facility
for at least 24 hours. If birth is at home, the first postnatal contact should be as early as
possible within 24 hours. At least three additional postnatal contacts are recommended for
all mothers and newborns on days 3 (48–72 hours), between days 7–14 after birth, and six
weeks after birth.
13
GUIDELINE: MANAGING POSSIBLE SERIOUS BACTERIAL INFECTION IN YOUNG INFANTS WHEN REFERRAL IS NOT FEASIBLE
Community health worker recommendations
The GDG made one recommendation for young infants 0–59 days old regarding home
visits by CHWs:
Recommendation 1: At home visits made as part of postnatal care, CHWs should
counsel families on recognition of danger signs, assess young infants for danger
signs and promote appropriate care seeking.
Strong recommendation based on moderate quality evidence
Remarks
The committee noted the following:
• CHWs should be provided with appropriate training, job aids, logistical support
and close monitoring and supervision in order to be able to identify danger signs.
Review Question 2: Fast breathing as a single sign of illness
The systematic review was based on the priority question: among neonates (0–28
days old) and young infants (0–59 days old)presenting with fast breathing as the only
sign of PSBI, whose families do not accept or cannot access referral care for treatment
(Population), are simpler antibiotic regimens delivered at outpatient and/or com­
munity level (Intervention) as effective as a combination of injectable penicillin and
gentamicin for at least seven days (Control) as measured by rates of mortality and
clinical deterioration (development of signs of severe infection or critical illness),
within two weeks of starting treatment (Outcome)?
Summary of evidence
Only one trial (the AFRINEST study [13]) was identified that was conducted among
children 0–59 days old with only fast breathing as a sign of PSBI (Population) whose
parents did not accept referral to hospital. The study, which is described in Grade
Table 2 (Annex 2), randomized 2333 infants (2196 in the per protocol analysis
reported in the GRADE tables) to either oral amoxicillin for seven days (Intervention)
or to intramuscular gentamicin plus intramuscular procaine penicillin (Control). The
study was a randomized controlled equivalency trial conducted in the Democratic
Republic of Congo, Kenya and Nigeria with equivalency defined a priori as having a
95% CI on the risk difference (RD) whose limits were between +/- 5.0%. Treatments
were evaluated based on the following four outcomes:
• Treatment failure in the first week after enrollment: The quality of evidence was
graded as low. A pooled analysis across the three sites showed the incidence
of treatment failure by day 8 after enrollment was 19.5% among those in the
intervention arm and 22.0% among the controls. The majority of failure was
due to persistence of fast breathing on day 5 (15.8% intervention versus 18.1%
control) There was no significant difference between the groups (RD -2.6%; 95%
CI: -6.0% to 0.8%), however, the lower limit of the CI was outside the equivalency
margin (i.e. below -5.0%). These results may indicate that the new or comparative
treatment could be better, but they are not conclusive. Age-specific subgroup
analysis between young infants 0–6 days of age (n=802) and 7–59 days of age
(n=1394) was carried out. In the 0–6 days of age group, treatment failure was
14
EVIDENCE AND RECOMMENDATIONS
22.8% (96/421) in the intervention arm and 24.7% (94/381) in the control arm.
There was no significant difference between the groups (RD -1.9%; 95% CI: -7.8%
to 4.0%). However, the lower limit of the CI was outside the equivalency margin.
• Mortality in first two weeks after enrollment: The quality of evidence was graded
as low. A pooled analysis across the three sites showed the incidence of death
by day 15 after enrollment was 0.2% (n=4) among those in the intervention arm
and 0.4% (n=4) among the controls. There was no significant difference between
the groups (RD -0.02%; 95% CI: -0.5% to 0.5%) and the RD along with its 95% CI lie
within the pre-specified equivalency margin indicating that the two regimens are
statistically equivalent. Age-specific subgroup analysis between young infants
0–6 days old (n=802) and 7–59 days old (n=1394) was carried out. In the 0–6 day
age group the incidence of death by day 15 after enrollment was 0.7% (n=3) among
those in the intervention arm and 0.1% (n=1) among the controls. The RD along
with the 95% CI did not lie within the pre-specified equivalency margin of +/- 0.5%
(RD 0.2%; 95% CI: -0.9% to 1.3%).
• Relapse and adherence: A pooled analysis across the three sites showed the
incidence of relapse during the second week after enrollment was 2.4% among the
914 intervention arm subjects who were initially well at day 8 and 2.2% among the
827 control arm subjects who were initially well at day 8. There was no significant
difference between the groups (RD 0.2%; 95% CI -1.2% to 1.6%), and the RD along
with its 95% CI lie within the pre-specified equivalency margin indicating that
the two regimens are statistically equivalent. Adherence was found to be better
in the intervention (98.5%) than in the control arm (90.7%) with adherence defined
as receiving 100% of doses of scheduled antibiotics on all seven days or by the
time of treatment failure.
Considerations for recommendation development
Balance of benefits versus harms: The GDG concluded that the benefits of the oral
regimen over the injectable regimen clearly outweigh potential harms. It also con­
cluded there was evidence for no difference in clinical efficacy (treatment failure),
serious adverse events or deaths when comparing seven days of oral amoxicillin to
seven days of intramuscular procaine penicillin and gentamicin. Finally it was noted
that patients given parenteral antibiotics had a lower rate of adherence with the
prescribed treatment compared to those given oral antibiotics.
Values and preferences: The GDG put a high value on the benefits of oral treatment.
The GDG felt that as the oral antibiotic regimen was equally as efficacious as the
injectable regimen, the oral regimen was preferred because it was simpler and
avoids injections. The GDG also considered that the oral regimen would be likely to
be more acceptable to families, be more accessible and have greater rates of treat­
ment completion.
Feasibility (including resource use considerations): The GDG concluded that oral
treatment is feasible. The members noted that oral amoxicillin is less expensive
than injectable antibiotics and could reduce the economic burden of treatment on
families and health systems. They also noted that oral amoxicillin is recommended
for treatment of pneumonia in children 2–59 months old and therefore should be
routinely available at most health facilities. However, the GDG did conclude that
15
GUIDELINE: MANAGING POSSIBLE SERIOUS BACTERIAL INFECTION IN YOUNG INFANTS WHEN REFERRAL IS NOT FEASIBLE
to ensure this regimen was effective, health care workers will need additional
training for skill building around diagnosis and treatment, regular supplies of oral
amoxicillin will be required, and the young infant component of current IMCI charts
will need revision.
Fast breathing recommendations
The GDG made two recommendations for children with fast breathing as the only sign
of illness whose families do not accept or cannot access referral, one for infants 0–6
days and one for infants 7–59 days of age.
Recommendation 2: Young infants 0–6 days old with fast breathing as the only
sign of illness should be referred to hospital. If families do not accept or cannot
access referral care, these infants should be treated with oral amoxicillin,
50 mg/kg per dose given twice daily for seven days, by an appropriately trained
health worker.
Strong recommendation, based on low quality evidence
Recommendation 3: Young infants 7–59 days old with fast breathing as the only
sign of illness should be treated with oral amoxicillin, 50 mg/kg per dose given
twice daily for seven days, by an appropriately trained health worker. These infants
do not need referral.
Strong recommendation, based on low quality evidence
Remarks
The GDG noted the following:
• These recommendations do not apply to infants with birth weight < 1500 g or
those who have been hospitalized for illness in the previous two weeks because
these interventions have not been tested in these populations. These infants
must be treated in a hospital.
• The health care worker should counsel the caregiver to return to the health care
worker immediately if the infant’s condition deteriorates.
• A mandatory follow-up by the health care worker should be made on the fourth
day of starting treatment (i.e. after completing three days of treatment).
• There is a higher risk of mortality in the first week of life compared to 7–59 days.
Further, the causes of fast breathing in the first week of life may not be due to
respiratory infection alone. The GDG therefore felt that 0–6 day old infants with
fast breathing as the only sign of illness should be referred to a hospital.
Review Question 3: Clinical severe infection1
The systematic review was based on the priority question: among neonates and young
infants (0–59 days old) with suspected/confirmed clinical severe infection whose families
do not accept or cannot access referral care for treatment (Population), can simpler
antibiotic regimens delivered at outpatient and/or community level (Intervention),
1
In a young infant (0–59 days old), at least one sign of severe infection i.e. movement only
when stimulated, not feeding well on observation, temperature greater than or equal to
38 °C or less than 35.5 °C or severe chest in-drawing.
16
EVIDENCE AND RECOMMENDATIONS
be as effective as a combination of injectable penicillin and gentamicin given for
at least seven days (Control) in achieving comparable rates of mortality, clinical
deterioration and persistence of signs of severe infection within two weeks of
starting treatment (Outcome)?
Summary of evidence
A total of four randomized controlled equivalency trials involving over 11 300
participants from Asia (Pakistan and Bangladesh) and Africa (Democratic Republic
of Congo, Kenya and Nigeria) have addressed this issue (10-12,23). They evaluated
five different antibiotic regimens, each in comparison to the standard of care of
seven days of intramuscular gentamicin plus seven days of intramuscular procaine
penicillin. The five regimens compared were:
1. intramuscular gentamicin for seven days + oral cotrimoxazole for seven days;
2. intramuscular ceftriaxone for seven days;
3. intramuscular gentamicin for seven days + oral amoxicillin for seven days;
4. intramuscular procaine penicillin + gentamicin for two days followed by oral
amoxicillin for five days;
5. intramuscular gentamicin for two days + oral amoxicillin for seven days.
Below is a summary of the evidence for each regimen compared to the reference
therapy. In cases where more than one study evaluated a regimen, a pooled analysis
was conducted using a fixed effects model. For the pooled analyses a simpler regimen
was considered ‘equivalent’ to the reference treatment if the 95% CI for the pooled
RD was within an equivalency margin of +/- 2.5% for treatment failure and +/- 0.5%
for death.
Regimen 1: Intramuscular gentamicin plus oral cotrimoxazole for seven days
Only one randomized trial from Pakistan (10) compared intramuscular gentamicin
plus oral cotrimoxazole for seven days (Intervention) to seven days of intramuscular
gentamicin plus seven days of intramuscular procaine penicillin (Control) in young
infants 0–59 days old whose parents did not accept referral to hospital (Population).
The study, which is described in Grade Table 3 (Annex 2), assigned 144 infants to
intramuscular gentamicin plus cotrimoxazole and 145 to intramuscular gentamicin
plus procaine penicillin (137 and 131 in per protocol analyses). Within the study
equivalency was defined a priori as a difference of ≤ 10.0%. Treatments were evaluated
based on the following outcomes:
• Treatment failure in the first week after enrollment: The quality of evidence was
graded as moderate. The failure rate was 9.9% in the standard of care arm and
18.2% in the intervention arm. The analysis showed significantly higher risk of
treatment failure in the first week with the simpler treatment of intramuscular
gentamicin plus oral cotrimoxazole for seven days (RD: 8.3%; 95% CI: 0.08% to
16.6%) compared to the standard of care (seven days of intramuscular gentamicin
plus seven days of intramuscular procaine penicillin).
• Mortality in the two weeks after enrollment: The quality of evidence was graded
as low. The mortality rate was 1.5% in the standard of care arm and 7.3% in the
intervention arm. The analysis showed an increased risk of mortality in the two
weeks after enrollment associated with the simpler treatment of intramuscular
gentamicin plus oral cotrimoxazole for seven days (RD: 5.8%; 95% CI: 0.9% to
17
GUIDELINE: MANAGING POSSIBLE SERIOUS BACTERIAL INFECTION IN YOUNG INFANTS WHEN REFERRAL IS NOT FEASIBLE
10.6%) versus the reference therapy (seven days of intramuscular gentamicin
plus seven days of intramuscular procaine penicillin).
• Relapse and adherence: Information on relapse could not be estimated from the
study. The rate of adherence was low overall and very similar between the two
groups (60.6% intervention and 58.7% control; RR: 1.03; 95% CI: 0.85–1.25).
Regimen 2: Intramuscular ceftriaxone for seven days
Only one randomized trial from Pakistan (10) compared intramuscular gentamicin
plus ceftriaxone for seven days (Intervention) to seven days of intramuscular
gentamicin plus seven days of intramuscular procaine penicillin (Control) in young
infants 0–59 days old whose parents did not accept referral to hospital (Population).
The study, which is described in Grade Table 4 (Annex 2), assigned 145 infants to
intramuscular ceftriaxone and 145 to intramuscular gentamicin plus procaine
penicillin (140 and 131 in per protocol analyses). Within the study, equivalency was
defined a priori as a difference of ≤ 10.0%. Treatments were evaluated based on the
following outcomes:
• Treatment failure in the first week after enrollment: The quality of evidence was
graded as very low. The failure rate was 9.9% in the standard of care arm and 15.7%
in the intervention arm. The analysis found no significant difference in the risk of
treatment failure between the two groups in the first week after enrollment (RD:
5.8%; 95% CI: -2.1% to 13.7%).
• Mortality in the two weeks after enrollment: The quality of evidence was graded
as very low. The mortality rate was 1.5% in the standard of care arm and 2.1% in
the intervention arm. The analysis found no significant difference in the risk of
mortality between the two groups in the two weeks after enrollment (RD: 0.6%;
95% CI: -2.6% to 3.8%).
• Relapse and adherence: Information on relapse could not be estimated from the
study. The rate of adherence was low overall but very similar between the two
groups (56.3% intervention versus 58.7% control; RR: 0.96; 95% CI: 0.78-1.17).
Regimen 3: Intramuscular gentamicin plus oral amoxicillin for seven days
Three randomized trials conducted in Bangladesh (12), Democratic Republic of
Congo, Kenya, Nigeria (11) and Pakistan (23) compared intramuscular gentamicin
plus oral amoxicillin for seven days (Intervention) to seven days of intramuscular
gentamicin plus seven days of intramuscular procaine penicillin (Control) in young
infants 0–59 days old whose parents did not accept referral to hospital (Population).
The three studies, which are described in Grade Table 5 (Annex 2), assigned a total
of 5054 (per protocol N=4729) infants to either gentamicin plus amoxicillin or to
intramuscular gentamicin plus procaine penicillin. Treatments were evaluated
based on the following outcomes:
• Treatment failure in the first week after enrollment: The quality of evidence was
graded as moderate. The rate of treatment failure was 8.1% in the intervention
arm and 9.9% in the control arm. Pooled analysis showed a significantly lower
rate of treatment failure in the first week after enrollment among those given
intramuscular gentamicin plus oral amoxicillin for seven days (RD: -1.8%; 95%
CI: -3.4% to -0.2%) compared to those given the standard of care (seven days of
intramuscular gentamicin plus seven days of intramuscular procaine penicillin).
18
EVIDENCE AND RECOMMENDATIONS
• Mortality in the two weeks after enrollment: The quality of evidence was graded
as moderate. The rate of mortality was 1.3% in the intervention arm and 1.6%
in the control arm. The pooled analysis showed no difference between the two
groups in the risk of mortality in the two weeks after enrollment although the
lower confidence limit was not within the pre-specified equivalency margin of
+/- 0.5% (RD: -0.3%; 95% CI: -1.0% to 0.4%).
• Relapse and adherence: There was no significant difference in the rate of relapse
between the groups, although relapse was less common among those given
intramuscular gentamicin plus oral amoxicillin for seven days (1.4% intervention
versus 1.9% control; RD -0.5%; 95% CI: -1.3% to 0.2%). The rate of adherence was
nearly identical (93.4% intervention versus 94.0% control) between the two groups
(RR: 0.99; 95% CI: 0.98-1.01).
Regimen 4: Intramuscular procaine penicillin plus gentamicin for two days
followed by oral amoxicillin for five days
Three randomized trials conducted in Bangladesh (12), Democratic Republic of Congo,
Kenya, Nigeria (11) and Pakistan (23) compared intramuscular procaine penicillin
plus gentamicin for two days followed by oral amoxicillin for five days (Intervention)
to seven days of intramuscular gentamicin plus seven days of intramuscular
procaine penicillin (Control) in young infants 0–59 days old whose parents did not
accept referral to hospital (Population). The three studies, which are described in
Grade Table 6 (Annex 2), assigned a total of 5066 (per protocol N=4775) infants to
either intramuscular procaine penicillin plus gentamicin for two days followed
by oral amoxicillin for five days or gentamicin plus penicillin (or a third regimen
described above). Treatments were evaluated based on the following outcomes:
• Treatment failure in the first week after enrollment: The quality of evidence was
graded as moderate. The rate of treatment failure was 9.5% in the intervention
arm and 9.9% in the control arm. The pooled analysis found no difference in the
incidence of treatment failure between the two groups in the first week after
enrollment (RD: -0.5%; 95% CI: -2.2% to 1.1%), and the results were within the prespecified equivalency margin of +/- 2.5%.
• Mortality in the two weeks after enrollment: The quality of evidence was graded
as moderate. The rate of mortality was 1.8% in the intervention arm and 1.7%
in the control arm. The pooled analysis found no significant difference in the
risk of mortality in the first two weeks after enrollment between the two groups,
although the results were not within the pre-specified equivalency margin of +/0.5% (RD: 0.1%; 95% CI: -0.6% to 0.9%).
• Relapse and adherence: There was a significant reduction in the rate of relapse
for those given intramuscular procaine penicillin plus gentamicin for two days
followed by oral amoxicillin for five days (0.8% intervention versus 1.9% control;
RD: -1.0%; 95% CI: -1.7% to -0.4%) versus those given the standard of care (seven
days of intramuscular gentamicin plus seven days of intramuscular procaine
penicillin). The rate of adherence was nearly identical between the two groups,
although only two studies reported adherence data (95.2% intervention versus
94.0% control; RR: 1.01; 95% CI: 1.0–1.03).
19
GUIDELINE: MANAGING POSSIBLE SERIOUS BACTERIAL INFECTION IN YOUNG INFANTS WHEN REFERRAL IS NOT FEASIBLE
Regimen 5: Intramuscular gentamicin for two days plus oral amoxicillin for seven
days
One multi-centre trial conducted in the Democratic Republic of Congo, Kenya and
Nigeria (11) compared intramuscular gentamicin for two days plus oral amoxicillin
for seven days (Intervention) to seven days of intramuscular gentamicin plus seven
days of intramuscular procaine penicillin (Control) in young infants 0–59 days old
whose parents did not accept referral to hospital (Population). The study, which is
described in Grade Table 7 (Annex 2), assigned a total of 1784 (per protocol N=1676)
infants to either intramuscular gentamicin for two days plus oral amoxicillin for
seven days or to intramuscular gentamicin plus procaine penicillin. Treatments
were evaluated based on the following outcomes:
• Treatment failure in the first week after enrollment: The quality of evidence was
graded as moderate. The rate of failure was 5.4% in the intervention arm and
8.1% in the control arm. The analysis showed that those given intramuscular
gentamicin for two days plus oral amoxicillin for seven days had significantly lower
incidence of treatment failure in the first week after enrollment compared to those
given seven days of intramuscular gentamicin plus seven days of intramuscular
procaine penicillin (RD: -2.7%; 95% CI: -5.1% to -0.3%).
• Mortality in the two weeks after enrollment: The quality of evidence was graded
as very low. The rate of mortality was 1.2% in the intervention arm and 1.3% in
the control arm. There was no significant difference in the risk of mortality in the
first two weeks after enrollment between the two groups, although the results
were not within the pre-specified equivalency margin of +/-0.5% (RD: -0.01%; 95%
CI: -1.2% to 0.9%).
• Relapse and adherence: There was no significant difference in the rate of relapse
between the groups, although relapse was less common among those given
intramuscular gentamicin for two days plus oral amoxicillin for seven days (0.8%
intervention versus 1.0% control; RD: -0.2%; 95% CI: -1.1% to 0.8%) compared
to those given seven days of intramuscular gentamicin plus seven days of
intramuscular procaine penicillin. The rate of adherence was nearly identical
between the two groups, but higher in the intervention arm (97.0% intervention
versus 92.5% control; RR: 1.05; 95% CI: 1.02-1.07).
Considerations for recommendation development
Balance of benefits versus harms
The GDG concluded the following:
• Oral cotrimoxazole plus intramuscular injectable gentamicin for seven days
(regimen 1) may be inferior to the reference regimen based on the higher rate
of treatment failure and clearly inferior to the reference regimen based on the
increased mortality.
• Intramuscular gentamicin plus oral amoxicillin for seven days (regimen 3) may
be superior to the reference treatment based on the reduced treatment failure
rate and non-inferior (equivalent or better) with respect to mortality.
• Intramuscular procaine penicillin plus gentamicin for two days followed by oral
amoxicillin for five days (regimen 4) is equivalent to the reference treatment
20
EVIDENCE AND RECOMMENDATIONS
based on the treatment failure rate and non-superior (equivalent or worse) based
on mortality.
• Intramuscular gentamicin for two days plus oral amoxicillin for seven days
(regimen 5) may be superior to the reference treatment due to lower treatment
failure, but is inconclusive with respect to mortality.
• None of the simpler regimens containing amoxicillin and gentamicin were
associated with an increased risk of serious adverse events or deaths compared
to the reference treatment.
Values and preferences: The GDG identified several benefits of simpler regimens that
would make them preferable over the current standard of care. The members noted
that compared to the reference treatment, which involves a total of 14 injections,
simpler antibiotic regimens containing amoxicillin and gentamicin would require
only two to seven injections. They concluded that families and health care providers
are likely to prefer fewer injections, while policy-makers and programme managers
are likely to give a high value to the simpler regimens that are easier to administer
in outpatient settings.
However, the GDG also noted some drawbacks to simpler regimens. Firstly, they
would require treatment to be provided by optimally trained and supervised
providers to avoid injection-related complications, and regular follow-up would
be needed for simpler antibiotic regimens to identify treatment failure including
clinical deterioration, need for change of antibiotic therapy, no improvement by
day 4 and referral for hospitalization. Secondly, in some countries there might be
regulatory barriers to the new regimens if certain cadres of health workers are not
permitted to administer antibiotic injections. Finally, extensive use of antimicrobial
agents for young infants may lead to AMR and therefore AMR surveillance would
be required. However, this risk is no greater than with the current WHO treatment
recommendations.
Feasibility (including resource use considerations): The GDG noted that simpler
antibiotic regimens are less expensive than the current standard treatment because
they require fewer injections and have fewer human resource requirements. In
addition oral amoxicillin is already recommended for treatment of pneumonia in
2–59 month old children, and therefore should be routinely available at most health
facilities. However, if simpler regimens were implemented, health care workers
would need training for skill building around diagnosis and treatment of PSBI, regular
supplies of injectable gentamicin and oral amoxicillin would need to be ensured and
changes in the young infant component of IMCI charts being used in countries would
need to be made.
Clinical severe infection recommendations
The GDG made one recommendation for young infants (0–59 days old) with clinical
severe infection who cannot access or refuse hospitalization, recommending two
treatment options.
Recommendation 4: Young infants (0–59 days) identified with clinical severe
infection whose families do not accept or cannot access hospital care should be
managed in outpatient settings by an appropriately trained health worker with
one of the two following regimens:
21
GUIDELINE: MANAGING POSSIBLE SERIOUS BACTERIAL INFECTION IN YOUNG INFANTS WHEN REFERRAL IS NOT FEASIBLE
Option 1: Intramuscular gentamicin 5–7.5 mg/kg (4) (for low-birth-weight infants
gentamicin 3–4 mg/kg) once daily for seven days and twice daily oral amoxicillin,
50 mg/kg per dose for seven days. Close follow-up is essential.
Strong recommendation based on moderate quality evidence
Option 2: Intramuscular gentamicin 5–7.5 mg/kg (4) (for low-birth-weight infants
gentamicin 3–4 mg/kg per day once daily) once daily for two days and twice daily
oral amoxicillin, 50 mg/kg per dose for seven days. Close follow-up is essential. A
careful assessment of the child on day 4 is mandatory for this option in order to
determine if the child is improving.
Strong recommendation based on low quality evidence
For option 2, the GRADE rating by the systematic review group was moderate quality
evidence for treatment failure and very low quality evidence for mortality. However,
the GDG reached a consensus agreement that the overall quality was low.
The GDG made a strong recommendation for option 2 despite low quality evidence.
The evidence was graded as low quality because data were available only from a
single, large, multi-country study from Africa while more evidence is available for
option 1 (graded as moderate quality). However there was no important difference in
the key outcomes. Option 1 is the preferred option, but where the health system does
not allow this to be implemented, option 2 could be considered. The GDG felt that
option 2 likely would be easier to deliver, have more equitable access, have higher
adherence, be more affordable and have similar effectiveness. It is expected that
individual countries will adapt the recommendations to suit the local social, cultural
and economic contexts. Countries are encouraged to hold key stakeholder discussions
to inform the decision-making on use and introduction of the recommendations into
national programmes.
Remarks
The GDG also noted the following:
• If any sign of critical illness appears at any time during treatment, the health care
worker should treat the young infant as a critically ill young infant.
• If any new sign of clinical severe infection (not present initially) appears after
48 hours of treatment, the health care worker should treat the young infant as a
critically ill young infant.
• If any sign of clinical severe infection is still present by day 8, the health care
worker should treat the young infant as a critically ill young infant.
• The health care worker should counsel families to return to the health care
worker immediately if the young infant’s condition worsens.
• At each contact (for both treatment and follow-up), the health care worker should
assess the baby for signs of clinical deterioration.
• Health care workers should promote exclusive breastfeeding, and the sick young
infant should be kept warm if the body temperature is low.
• The GDG strongly encourages governments to set up country level surveillance
for AMR to guide future antibiotic decisions.
22
EVIDENCE AND RECOMMENDATIONS
Review Question 4: Critical illness1
The systematic review was based on the priority question: among neonates and
young infants (0–59 days old) with signs of critical illness, whose families do not
accept or cannot access referral care for treatment (Population), are simpler antibiotic
regimens delivered at outpatient and/or community level (Intervention), as effective
as a combination of injectable penicillin and gentamicin for at least seven days
(Control) as measured by rates of mortality, clinical deterioration and persistence of
signs of severe infection, within two weeks of starting treatment (Outcome)?
Summary of evidence
There were no comparative trials on which to base this recommendation.
Critical illness recommendations
The GDG made one recommendation for young infants 0–59 days old who have any
sign of critical illness:
Recommendation 5: Young infants 0–59 days old who have any sign of critical
illness (at presentation or developed during treatment of clinical severe infection)
should be hospitalized after pre-referral treatment with antibiotics.2
Strong recommendation based on very low quality evidence (standard of
care)
Remarks
Although there are no comparative trials available showing the relative efficacy
and safety, in cases where hospitalization is not possible at all, critically ill children
should be given one of the following treatment regimens until hospitalization
becomes possible (up to seven days):
1. twice daily intramuscular ampicillin and once daily intramuscular gentamicin;
2. once daily intramuscular ceftriaxone with or without once daily intramuscular
gentamicin;
3. twice daily intramuscular benzyl penicillin and once daily intramuscular genta­
micin;
4. once daily intramuscular procaine penicillin and once daily intramuscular genta­
micin.
1
In a sick young infant, presence of any of the following signs: unconscious, convulsions,
unable to feed at all, apnoea, unable to cry, cyanosis, bulging fontanelle, major congenital
malformations inhibiting oral antibiotic intake, active bleeding requiring transfusion,
surgical conditions needing hospital referral, persistent vomiting (defined as vomiting
following three attempts to feed the infant within 30 minutes and the infant vomits after
every attempt).
2
Give first dose of both ampicillin (50 mg/kg per dose) or benzyl penicillin (50 000 units/kg per
dose) and gentamicin (5–7.5 mg/kg per dose) intramuscularly.
23
GUIDELINE: MANAGING POSSIBLE SERIOUS BACTERIAL INFECTION IN YOUNG INFANTS WHEN REFERRAL IS NOT FEASIBLE
Dissemination, implementation
and monitoring of guideline
Dissemination
The recommendations in this guideline will be disseminated through a broad
network of international partners, including WHO country and regional offices,
ministries of health, WHO collaborating centres, other United Nations agencies and
non-governmental organizations. They will also be published on the WHO website.
Strategic dissemination to key stakeholders will ensure that the guideline reaches
the users most likely to benefit from it.
Adaptation and implementation
The first steps in implementation will be to revise all WHO publications that deal
with these conditions, and/or ensure their inclusion in other relevant documents.
These include the materials for the Integrated Management of Childhood Illness (3) and
The Pocket Book for Hospital Care of Children (4).
The successful introduction of evidence-based policies related to management of
sick young infants into national programmes and health care services depends
on well-planned and participatory consensus-driven processes of adaptation and
implementation. These processes may include the development or revision of
existing national guidelines or protocols based on this document.
It is expected that individual countries will adapt the recommendations to suit the
local social, cultural and economic contexts. Countries will be encouraged to hold key
stakeholder discussions to inform decision-making on the use and introduction of
the recommendations into national programmes. The recommendations contained
in the present guideline should be adapted into locally-appropriate documents to
meet the specific needs of each country and health service. For management of
‘clinical severe infection’ in young infants where referral is not possible, the preferred
option is the first one using seven injections. However, where the health system does
not allow this to be implemented, the second option could be considered. MCA will
explore using a framework for assisting policy-makers in adapting this guideline,
such as the DECIDE framework (24).
An enabling environment should be created for the use of these recommendations,
including changes in the behaviour of health care practitioners to enable the use of
evidence-based practices. Local professional societies may play important roles in
this process, and an all-inclusive and participatory process should be encouraged.
MCA has substantial experience of introduction of WHO guidelines and tools into
national programmes.
The drugs recommended in this document are on the WHO Model List of Essential
Medicines (25). Essential medicines are intended to be available within the context
of functioning health systems at all times in adequate amounts, in the appropriate
dosage forms, with assured quality, and at a price the individual and the community
24
DISSEMINATION, IMPLEMENTATION AND MONITORING OF GUIDELINE
can afford. The Model List is a guide for the development of national and institutional
essential medicine lists.
Within this context, programme managers will need to ensure that adequate
quantities of required drugs in the recommended dosages are available to health
workers. These drugs would normally be provided through existing health system
supply chains.
Relevant to any review of national drug lists and policy, WHO’s 2014 report on global
surveillance of AMR reveals that antibiotic resistance is a serious and growing
problem across the world (26). Countries should strengthen national plans to tackle
AMR.
Monitoring and evaluation of guideline implementation
Monitoring and evaluation should be built into the implementation process, in order
to provide important lessons for uptake and further implementation. With regard to
monitoring and evaluation of their impact on quality of care, priority should be given
to the strong recommendations.
The implementation of this guideline should involve national child health pro­
grammes collecting and reporting data on the management of sick young infants
whose families do not accept or cannot access referral care. Putting this into practice
may require a review of existing patient monitoring systems, including reporting
tools, to ensure that the conditions are adequately addressed.
Key areas that may require monitoring include:
• diagnosis of sick young infants with suspected sepsis;
• treatment of sick young infants with suspected sepsis at first level health facilities;
• response to treatment;
• risk of side effects especially from gentamicin toxicity;
• service delivery (including the need for metrics to track coverage and quality of
care and adherence to treatment protocols);
• support systems, including supplies and logistics, and supervision;
• community perceptions and acceptance.
Global and country level efforts are underway under Every Newborn: an action plan
to end preventable deaths (27,28) through Metrics Task Teams to agree upon critical
indicators for management of neonatal sepsis. The monitoring and evaluation
strategy will endeavour to ensure that the existing patient monitoring tools at health
facilities and communities will contain information on recognition and manage­
ment of sick young infants. However, the data could be collected periodically through
special surveys or programme reviews.
MCA will also monitor implementation of this guideline using indicators such as
the number of requests from countries for assistance in implementation as well as
requests to WHO headquarters and regional offices for monitoring and evaluation
in countries applying the guideline. MCA will work with the regional offices to
monitor the number of countries implementing this guideline. Additionally, MCA
will monitor the number of downloads of the guideline document from the WHO,
partners’ and other stakeholders’ websites, as well as the number of hard copies of
the guidance requested and distributed through the WHO document centre.
25
GUIDELINE: MANAGING POSSIBLE SERIOUS BACTERIAL INFECTION IN YOUNG INFANTS WHEN REFERRAL IS NOT FEASIBLE
Implications for future research
Before countries can use this guideline for policy adoption and implementation at
scale, more work is needed to address these as well as operational issues. Important
consideration should also be given to tracking possible harm, for example at the
individual level there is a risk of increased gentamicin toxicity (e.g. ototoxicity
and nephrotoxicity) particularly if blood levels are not being tracked. Innovative
approaches are needed to achieve this safety tracking at lower levels of the health
system. At a population level, innovative and lower costs methods are needed for
tracking of potential changes in AMR.
Facilitating policy adoption and putting in place an enabling environment for
implementation will require a dialogue with policy-makers and other stakeholders at
country level. Programme managers will require technical support for development
and implementation of operational plans in programme settings from experts
with experience delivering these interventions. If policy dialogue and small-scale
demonstration projects are not supported at the country level, it is less likely that
this guideline will be adopted as policy and scaled-up.
To achieve scale-up WHO and UNICEF will jump-start the process through country
level policy dialogue to facilitate small-scale demonstration projects for simplified
management of sick young infants with suspected sepsis where referral is not
possible. This will facilitate future scale-up of these lifesaving interventions. The
approach will have three components.
• Policy dialogue and orientation meetings held at the national level with ministries
of health and other stakeholders at the national and subnational levels to discuss
limited policy adoption to set up demonstration sites in a few health facilities.
• Demonstration sites established in a few countries to demonstrate feasibility
of delivering simplified antibiotic regimens to young infants with suspected
serious bacterial infection where families do not accept or cannot access
referral care. The innovation of this approach will be in the careful assessments
and measurements of the logistical requirements for programme uptake and
expansion. MCA will document the human resource requirements, the steps
needed to make the supply chain for essential commodities work consistently,
the level of supervision required and the ways in which that level of supervision
can be produced given the limits of current supervisory systems in many areas.
This implementation research should provide four country-specific assessments
of the true barriers to implementation and success, and will be able to offer some
grounded recommendations on overcoming those barriers. We aim to monitor
the quality of programme implementation through outcomes such as whether
80% of young infants receive “adequate quality” treatment. More details are out­
side the scope of this document.
• Develop a partnership between demonstration sites and programme managers
to provide technical assistance to initiate a pilot in other health facilities.
26
DISSEMINATION, IMPLEMENTATION AND MONITORING OF GUIDELINE
Plans for updating the guideline
This guideline will be reviewed and updated in January 2019. At that time the WHO
Steering Committee will constitute a GDG to review the literature and update the
recommendations as needed. In the interim, the WHO Steering Committee will
continue to monitor any new studies, interim research results or reports of adverse
events associated with this policy implementation. If relevant information becomes
available that indicates urgent changes to the recommendations before January
2019, a GDG will be constituted at that time.
27
GUIDELINE: MANAGING POSSIBLE SERIOUS BACTERIAL INFECTION IN YOUNG INFANTS WHEN REFERRAL IS NOT FEASIBLE
References
1 Walker N, Fischer-Walker C, Bryce J, Bahl R, Cousens S, CHERG Review Groups
on Intervention Effects. Standards for CHERG reviews of intervention effects on
child survival. Int J Epidemiol. 2010; 39: 21–31.
2 World Health Organization. WHO recommendations on postnatal care of
the mother and newborn. Geneva: WHO; 2013 (http://apps.who.int/iris/
bitstream/10665/97603/1/9789241506649_eng.pdf, accessed 15 May 2015).
3 World Health Organization. Integrated Management of Childhood Illness
(IMCI), chart booklet. Geneva: WHO; 2014 (http://apps.who.int/iris/
bitstream/10665/104772/16/9789241506823_Chartbook_eng.pdf, accessed 15
May 2015).
4 World Health Organization. Pocket book of hospital care for children, second
edition. Guidelines for the management of common childhood illnesses.
Geneva: WHO; 2013 (http://www.who.int/maternal_child_adolescent/
documents/child_hospital_care/en/, accessed 8 July 2015).
5 UNICEF, WHO, World Bank, United Nations Population Division. Levels and
trends in child mortality: report 2014. Estimates developed by the UN Interagency Group for Child Mortality Estimation. New York: UNICEF; 2014.
6 Liu L, Oza S, Hogan D, Perin J, Rudan I, Lawn JE et al. Global, regional, and
national causes of child mortality in 2000–13, with projections to inform post2015 priorities: an updated systematic analysis. Lancet. 2014; 385: 430–40.
7 Bhandari N, Bahl R, Bhatnagar V, Bhan MK. Treating sick young infants in urban
slum setting. Lancet. 1996; 347: 1774–5.
8 Bang AT, Bang RA, Baitule SB, Reddy MH, Deshmukh MD. Effect of home-based
neonatal care and management of sepsis on neonatal mortality: field trial in
rural India. Lancet. 1999; 354: 1955–61.
9 Baqui AH, El-Arifeen S, Darmstadt GL, Ahmed S, Williams EK, Seraji HR et al.
Effect of community-based newborn-care intervention package implemented
through two service-delivery strategies in Sylhet district, Bangladesh: a clusterrandomised controlled trial. Lancet. 2008; 371: 1936–44.
10 Zaidi AKM, Tikmani SS, Warraich HJ, Darmstadt GL, Bhutta ZA, Sultana S et
al. Community-based treatment of serious bacterial infections in newborns
and young infants: a randomized controlled trial assessing three antibiotic
regimens. Pediatr Infect Dis J. 2012; 31: 667–72.
11 African Neonatal Sepsis Trial (AFRINEST) Group, Tshefu A, Lokangaka A,
Ngaima S, Engman C, Esamai F et al. Simplified antibiotic regimens compared
with injectable procaine benzylpenicillin plus gentamicin for treatment of
neonates and young Infants with clinical signs of severe infection when referral
is not possible: a randomized equivalence trial. Lancet. 2015; 385: 1767–76.
28
REFERENCES
12 Baqui A, Saha S, Ahmed A, Shahidulla M, Quasem I, Roth DE et al. Safety and
efficacy of alternative antibiotic regimens compared with 7 day injectable
procaine benzylpenicillin and gentamicin for outpatient treatment of neonates
and young infants with clinical signs of severe infection when referral is not
possible: a randomised open-label, equivalence trial. Lancet Glob Heal. 2015; 3:
e279–87.
13 African Neonatal Sepsis Trial (AFRINEST) Group, Tshefu A, Lokangaka A,
Ngaima S, Engmann C, Esamai F et al. Oral amoxicillin compared with injectable
procaine benzylpenicillin plus gentamicin for treatment of neonates and young
infants with fast breathing when referral is not possible: a randomised, openlabel, equivalence trial. Lancet. 2015; 385: 1758–66.
14 Kumar V, Mohanty S, Kumar A, Misra RP, Santosham M, Awasthi S et al. Effect
of community-based behaviour change management on neonatal mortality in
Shivgarh, Uttar Pradesh, India: a cluster-randomised controlled trial. Lancet.
2008; 372: 1151–62.
15 World Health Organization. Case management of acute respiratory infections in
children in developing countries. Geneva: Switzerland; 1985.
16 World Health Organization. Handbook for guideline development, second
edition. Geneva: Switzerland; 2014 (http://www.who.int/kms/handbook_2nd_
ed.pdf, accessed 15 May 2015).
17 GRADE Working Group. Grading guide – grading the quality of evidence and
strength of recommendations (http://www.uptodate.com/home/grading-guide,
accessed on 15 May 2105).
18 Bashour HN, Kharouf MH, Abdulsalam AA, El Asmar K, Tabbaa MA, Cheikha
SA. Effect of postnatal home visits on maternal/infant outcomes in Syria: a
randomized controlled trial. Public Health Nurs. 2008; 25: 115–25.
19 Bhutta ZA, Soofi S, Cousens S, Mohammed S, Memon ZA, Ali I et al.
Improvement of perinatal and newborn care in rural Pakistan through
community-based strategies: a cluster-randomised effectiveness trial. Lancet.
2011; 377: 403–12.
20 Darmstadt GL, Choi Y, Arifeen SE, Bari S, Rahman SM, Mannan I et al. Evaluation
of a cluster-randomized controlled trial of a package of community-based
maternal and newborn interventions in Mirzapur, Bangladesh. PLoS One. 2010;
5: e9696.
21 Kirkwood BR, Manu A, ten Asbroek AHA, Soremekun S, Weobong B, Gyan T et al.
Effect of the Newhints home-visits intervention on neonatal mortality rate and
care practices in Ghana: a cluster randomised controlled trial. Lancet. 2013; 381:
2184–92.
22 Mazumder S, Taneja S, Bahl R, Mohan P, Strand TA, Sommerfelt H et al. Effect of
implementation of Integrated Management of Neonatal and Childhood Illness
programme on treatment seeking practices for morbidities in infants: cluster
randomised trial. BMJ. 2014; 349: g4988.
29
GUIDELINE: MANAGING POSSIBLE SERIOUS BACTERIAL INFECTION IN YOUNG INFANTS WHEN REFERRAL IS NOT FEASIBLE
23 Mir F, Nisar I, Tikmani SS, Baloch B, Shakoor S, Jehan F et al. Simplified antibiotic
regimens for the management of sepsis in young infants in first-level facilities:
randomized controlled trial Pakistan. Under review, 2015.
24 Developing and evaluating communication strategies to support informed
decisions and practice based on evidence (DECIDE) (http://www.decidecollaboration.eu/, accessed 1 July, 2015).
25 World Health Organization. Model list of essential medicines for children,
fourth edition. Geneva: Switzerland; 2013 (http://apps.who.int/iris/
bitstream/10665/93143/1/EMLc_4_eng.pdf?ua=1, accessed 15 may 2015).
26 World Health Organization. Antimicrobial resistance: global report on
surveillance 2014. Geneva: Switzerland; 2014 (http://apps.who.int/iris/
bitstream/10665/112642/1/9789241564748_eng.pdf, accessed 15 May 2015).
27 World Health Organization, UNICEF. Every newborn: an action plan to end
preventable deaths. Geneva: Switzerland; 2014 (http://www.who.int/maternal_
child_adolescent/topics/newborn/enap_consultation/en/, accessed 15 May
2015).
28 Mason E, McDougall L, Lawn J, Gupta A, Claesen M, Pillay Y et al. From evidence
to action to deliver a healthy start for the next generation. Lancet. 2014; 384:
455–67.
30
ANNEX 1 DECLARATION OF INTERESTS
ANNEX 1
Declaration of interests
The Declaration of Interest for WHO Experts forms of all the invited participants
and temporary advisers for the guideline meetings were reviewed and assessed.
Seven participants declared having interests that could be perceived to be a conflict
of interest. The experts were asked to confirm their declarations of interest verbally
at the beginning of the meeting.
1. Dr W Carlo, University of Alabama at Birmingham, USA, declared having
current investments in a commercial entity, MEDNAX. MEDNAX, Inc., is a
national medical group that comprises the USA’s leading providers of neonatal,
anaesthesia, maternal-fetal and paediatric physician subspecialty services.
2. Dr C Garg, Consultant, New Delhi, India, declared having received consulting
fees from WHO.
3. Dr P Hibberd, Massachusetts General Hospital, Boston, USA, declared having
received a research grant from Saving Lives at Birth.
4. Dr E Molyneux, College of Medicine, Blantyre, Malawi, declared having received
a research grant from Saving Lives at Birth and a grant from GSK-Save the
Children Partnership to introduce bubble continuous positive airway pressure
training.
5. Dr K Mulholland, Murdoch Children’s’ Research Institute, Melbourne, Australia,
declared having been for two years the Chair of the Technical Steering Committee
for the Simplified Antibiotic Treatment trials in Bangladesh and Pakistan.
6. Dr V Paul, All India Institute of Medical Sciences, New Delhi, India, declared
being an academic who takes public positions on child health issues including
treatment of acute respiratory infections.
7. Dr D Vivio, United States Agency for International Development, Washington
DC, USA, declared that the Agency contributed towards her travel costs in
connection with this meeting.
External resource persons were invited to the meeting as observers and to provide
technical input, but they did not participate in the decision-making processes. All
declarations of interests were reviewed, and none were found to be a conflict of
interest for the experts invited to this guideline development meeting. None of the
above-mentioned persons chaired the meeting.
31
With intervention
With comparator
537 per 1000
RR 1.35
(1.15 to 1.58)
Relative effect
(95% CI)
9158
(6 studies)
No. of participants
(studies)
a
Blinding was not achievable given the nature of intervention amongst the two groups. The group allocated was easily identifiable by interview with patients.
Home visits by CHWs/
no home visits
725 per 1000
(618 to 849)
Corresponding risk
Assumed risk
Illustrative comparative risks (95% CI)
Improved care seeking from health facility (14,18-22)
Outcomes
Intervention
and comparison
intervention
⊕⊕⊕
moderatea
Quality of the
evidence
(GRADE)
GRADE TABLE 1. Home visits by CHWs for improved care seeking from health facilities in newborns and young infants
GRADE summary of findings tables
ANNEX 2
see footnotes
Comments
GUIDELINE: MANAGING POSSIBLE SERIOUS BACTERIAL INFECTION IN YOUNG INFANTS WHEN REFERRAL IS NOT FEASIBLE
32
33
Risk of bias
Inconsistency
Indirectness
Imprecision
randomized
trial (13)
not serious
single study
1
randomized
trial (13)
not serious
single study
Deaths in first two weeks after enrollment
1
not serious
not serious
seriousc
seriousb
b
Intramuscular
95% CI around the pooled estimate includes 1) null effect and 2) appreciable harm/benefit.
c
Few events.
a
Study design
Oral
amoxicillin
for 7 days
IMa
penicillin
plus IM
gentamicin
for 7 days
No. of patients
Risk
difference
(95% CI)
Effect of the
intervention
vs control
Quality
Importance
Interpretation
4/1135
(0.4%)
221/1135
(19.5%)
4/1061
(0.4%)
234/1061
(22.0%)
0 fewer per
1000
(from 5 more
to 5 fewer)
RD -0.02%
(-0.5% to
0.5%)
26 fewer per
1000 (from
8 more to 60
fewer)
RD -2.6%
(-6.0% to
0.8%)
⊕⊕
LOW
⊕⊕
LOW
CRITICAL
CRITICAL
New regimen is
EQUIVALENT to
reference treatment
The lower limit of CI
crosses equivalence
margins, which
might indicate
that the new or
comparative
treatment may be
better, but is not
conclusive
Treatment failure by day 8 after enrollment (assessed with: death, clinical deterioration, change of antibiotics, no improvement by day 4, etc., in the first week of
enrollment)
No. of
studies
Quality assessment
GRADE TABLE 2: Treatment of fast breathing pneumonia: oral amoxicillin for seven days versus intramuscular penicillin plus intramuscular
gentamicin for seven days
ANNEX 2. GRADE SUMMARY OF FINDINGS TABLES
34
Risk of bias
Inconsistency
randomized
trial (10)
seriousa
single study
1
randomized
trial (10)
not seriousb
single study
Deaths in first two weeks after enrollment
1
c
b
Neither intervention nor outcome assessment was blinded.
Outcome assessment not blinded but outcome is ‘objective’.
Few events.
a
Study design
Indirectness
Imprecision
IM gentamicin
plus oral
cotrimoxazole
for 7 days
IM penicillin
plus IM
gentamicin
for 7 days
No. of patients
Risk
difference
(95% CI)
Effect of the
intervention
vs control
Quality
Importance
not serious
not serious
seriousc
not serious
10/137 (7.3%)
25/137
(18.2%)
2/131
(1.5%)
13/131
(9.9%)
58 more per
1000
(from 9
more to 106
more)
RD 5.8%
(0.9% to
10.6%)
83 more per
1000
(from 8
more to 166
more)
RD 8.3%
(0.08% to
16.6%)
⊕⊕
LOW
⊕⊕⊕
MODERATE
CRITICAL
CRITICAL
Treatment failure by day 8 after enrollment (assessed with: death, clinical deterioration, change of antibiotics, no improvement by day 4, etc.,
in the first week of enrollment)
No. of
studies
Quality assessment
New regimen
is not
equivalent –
it is INFERIOR
to reference
treatment
New regimen
is not
equivalent –
it is INFERIOR
to reference
treatment
Interpretation
GRADE TABLE 3: Treatment of PSBI: intramuscular gentamicin plus oral cotrimoxazole for seven days versus intramuscular penicillin plus
intramuscular gentamicin for seven days
GUIDELINE: MANAGING POSSIBLE SERIOUS BACTERIAL INFECTION IN YOUNG INFANTS WHEN REFERRAL IS NOT FEASIBLE
35
b
Risk of
bias
Inconsistency
1
randomized
trial (10)
seriousa
single study
Indirectness
Imprecision
randomized
trial (10)
not
seriousc
single study
not serious
not serious
very
seriousb,d
serious b
Neither intervention nor outcome assessment was blinded.
95% CI around the pooled estimate includes 1) null effect and 2) appreciable harm/benefit.
Outcome assessment not blinded but outcome is ‘objective’.
Few events.
1
Deaths in first two weeks after enrollment
c
d
a
Study
design
IM ceftriaxone
for 7 days
IM penicillin
plus IM
gentamicin for
7 days
No. of patients
Risk
difference
(95% CI)
Effect of the
intervention
vs control
Quality
Importance
Interpretation
3/140
(2.1%)
22/140
(15.7%)
2/131
(1.5%)
13/131
(9.9%)
6 more per
1000
(from 26
fewer to 38
more)
RD 0.6%
(-2.6% to
3.8%)
58 more per
1000
(from 21
fewer to 137
more)
RD 5.8%
(-2.1% to
13.7%)
⊕
VERY LOW
⊕
VERY LOW
CRITICAL
CRITICAL
Wide 95% CI
– equivalence
cannot be
commented
upon
The upper limit
of CI crosses
equivalence
margins, which
may indicate
that the new or
comparative
treatment may
be inferior, but
is not conclusive
Treatment failure by day 8 after enrollment (assessed with: death, clinical deterioration, change of antibiotics, no improvement by day 4, etc., in the first week of
enrollment)
No. of
studies
Quality assessment
GRADE TABLE 4: Treatment of PSBI: intramuscular ceftriaxone for seven days versus intramuscular penicillin plus intramuscular gentamicin
for seven days
ANNEX 2. GRADE SUMMARY OF FINDINGS TABLES
36
b
Risk of
bias
Inconsistency
3
randomized
trials
(11,12,23)
seriousa
not serious
Indirectness
Imprecision
IM gentamicin
plus oral
amoxicillin
for 7 days
IM penicillin
plus IM
gentamicin
for 7 days
Risk
difference
(95% CI)
Effect of the
intervention
vs control
Quality
Importance
Interpretation
randomized
trials
(11,12,23)
not
seriousb
not serious
not serious
not serious
seriousc,d
not serious
31/2417
(1.3%)
192/2359
(8.1%)
39/2436
(1.6%)
235/2370
(9.9%)
3 fewer per
1000
(from 4
more to 10
fewer)
RD -0.3%
(-1.0% to
0.4%)
18 fewer per
1000
(from 2
fewer to 34
fewer)
RD -1.8%
(-3.4% to
-0.2%)
⊕⊕⊕
MODERATE
⊕⊕⊕
MODERATE
CRITICAL
CRITICAL
The lower limit
of CI crosses
equivalence
margins, which
may indicate
that the new or
comparative
treatment might
be better, but is
not conclusive
New regimen is
not equivalent –
it is SUPERIOR
to reference
treatment
Unclear risk of measurement bias (in two out of three studies, physicians masked to group allocation only confirmed the cases referred to them; they did not measure the outcomes in all enrolled infants).
Unclear risk of measurement bias in two studies but outcome is ‘objective’ .
95% CI around the pooled estimate includes 1) no effect and 2) appreciable harm/benefit.
Few events but sample size > 2000 per group.
3
Deaths in first two weeks after enrollment
c
d
a
Study
design
No. of patients
Treatment failure by day 8 after enrollment (assessed with: death, clinical deterioration, change of antibiotics, no improvement by day 4, etc., in the first week of
enrollment)
No. of
studies
Quality assessment
GRADE TABLE 5: Treatment of PSBI: intramuscular gentamicin plus oral amoxicillin for seven days versus intramuscular penicillin plus
intramuscular gentamicin for seven days
GUIDELINE: MANAGING POSSIBLE SERIOUS BACTERIAL INFECTION IN YOUNG INFANTS WHEN REFERRAL IS NOT FEASIBLE
37
b
Risk of bias
Inconsistency
3
randomized
trials
(11,12,23)
seriousa
not serious
Indirectness
Imprecision
IM penicillin
plus IM
gentamicin
for 7 days
Risk
difference
(95% CI)
Effect of the
intervention
vs control
Quality
Importance
Interpretation
randomized
trials
(11,12,23)
not
seriousb
not serious
not serious
not serious
seriousc,d
not serious
43/2405
(1.8%)
228/2405
(9.5%)
39/2307
(1.7%)
235/307
(9.9%)
1 more per
1000 (from
6 more to 9
fewer)
RD 0.1%
(-0.6% to
0.9%)
5 fewer per
1000 (from
11 more to 22
fewer)
RD -0.5%
(-2.2% to
1.1%)
⊕⊕⊕
MODERATE
⊕⊕⊕
MODERATE
CRITICAL
CRITICAL
Wide 95% CI
– equivalence
cannot be
commented
upon
New regimen
is EQUIVALENT
to reference
treatment
Unclear risk of measurement bias (in two out of three studies, physicians masked to group allocation only confirmed the cases referred to them; they did not measure the outcomes in all enrolled infants).
Unclear risk of measurement bias in two studies but outcome is ‘objective’.
95% CI around the pooled estimate includes 1) no effect and 2) appreciable harm/benefit.
Few events but sample size > 2000 per group.
3
Deaths in first two weeks after enrollment
c
d
a
Study design
IM penicillin +
gentamicin for 2
days followed by
oral amoxicillin
for 5 days
No. of patients
Treatment failure by day 8 after enrollment (assessed with: death, clinical deterioration, change of antibiotics, no improvement by day 4, etc., in the first week of
enrollment)
No. of
studies
Quality assessment
GRADE TABLE 6: Treatment of PSBI: intramuscular procaine penicillin plus gentamicin for two days followed by oral amoxicillin for five days
versus intramuscular gentamicin plus intramuscular procaine penicillin for seven days (reference treatment)
ANNEX 2. GRADE SUMMARY OF FINDINGS TABLES
38
Risk of bias
Inconsistency
Indirectness
Imprecision
randomized not serious
trial (11)
single study
1
randomized not serious
trial (11)
single study
Deaths in first two weeks after enrollment
1
not serious
not serious
very
seriousa,b
not serious
b
95% CI around the pooled estimate includes 1) no effect and 2) appreciable harm/benefit.
Few events.
a
Study
design
IM gentamicin
for 2 days and
oral amoxicillin
for 7 days
IM penicillin
plus IM
gentamicin
for 7 days
No. of patients
Risk
difference
(95% CI)
Effect of the
intervention
vs control
Quality
Importance
Interpretation
11/890
(1.2%)
46/848
(5.4%)
12/894
(1.3%)
67/828
(8.1%)
0 fewer per
1000
(from 9 more
to 12 fewer)
RD -0.01%
(-1.2% to
0.9%)
27 fewer per
1000
(from 3 fewer
to 51 fewer)
RD -2.7%
(-5.1% to
-0.3%)
⊕
VERY LOW
⊕⊕⊕
MODERATE
CRITICAL
CRITICAL
Wide 95% CI
– equivalence
cannot be
commented
upon
New regimen
is not
equivalent –
it is SUPERIOR
to reference
treatment
Treatment failure by day 8 after enrollment (assessed with: death, clinical deterioration, change of antibiotics, no improvement by day 4, etc., in the first week of
enrollment )
No. of
studies
Quality assessment
GRADE TABLE 7: Treatment of PSBI: intramuscular gentamicin for two days plus oral amoxicillin for seven days versus intramuscular
gentamicin plus intramuscular procaine penicillin (reference treatment)
GUIDELINE: MANAGING POSSIBLE SERIOUS BACTERIAL INFECTION IN YOUNG INFANTS WHEN REFERRAL IS NOT FEASIBLE
For more information please contact:
Department of Maternal, Newborn, Child and Adolescent Health (MCA)
World Health Organization
20 Avenue Appia, 1211 Geneva 27, Switzerland
Tel: + 41 22 791 32 81; Fax: + 41 22 791 48 53
E-mail: [email protected]
Website: http://www.who.int/maternal_child_adolescent
ISBN 978 92 4 150926 8